Brain glioma tumors, unequivocally glioblastoma multiforme (GBM), are among the most incurable forms of cancer. Such tumors are thought to arise from brain neural stem cells and progenitors that have undergone transformation into neoplasias. Neoplastic stem cells are thought to contribute to the recurrence of the glioma after chemotherapy and surgical resection. We found that activating transcription factor 5 (ATF5) is highly expressed in neural stem/progenitor cells, including cancer stem cells and GBMs. Blocking ATF5 function by dominant negative ATF5 (d/n-ATF5) or siRNA-ATF5 promotes apoptosis of glioma tumor cells, but not of non- neoplastic cells, both in vitro and in vivo. Currently, we synthesize recombinant cell penetrant d/n-ATF5 peptide that crosses the blood brain barrier, and enters into glioma cells, promoting their rapid death. The long-term objectives of our study will be to further test, as well as determine the apparent therapeutic index, of cell penetrant d/n-ATF5 toward tumor regression or full eradication in a pre-clinical approach by using different glioblastoma mouse models. Our specific aims will be to 1) Define the most effective dosing schedule for delivery of the cell penetrant d/n-ATF5 that does not harm normal tissues; test efficacy in mouse glioma/glioblastoma models by creating brain tumors through de novo transformation of progenitors and by human GBM xenografts; determine whether treatment with the peptide can bring about long-term eradication of glioblastomas; and, in the case that tumors reappear after initial treatment, whether they can again be caused to regress by application of the d/n peptide. 2) To define the responsible molecular mechanistic pathways that mediate apoptotic actions of d/n-ATF5 in glioblastoma cells. To gain insight on these pathways will enlighten how neoplasm relies on ATF5 for survival that is not observed in non-transformed cells. Knowledge of the mechanistic routes will aid in predicting and circumventing off-target effects and will explain as well as promote avoidance of potential tumor resistance toward d/n-ATF5 therapy. Finally, synergy of d/n- ATF5 with other currently employed glioblastoma therapies will be more adequately addressed with awareness of such pathways.